=== johnlage_partyha is now known as johnlage === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === la_juyis` is now known as la_juyis === qwebirc166433 is now known as slickymasterWork [16:07] elopio, hey i have an autopilot test failing because the item it's looking for is off the end of the list - what's the best way to handle that? i remember something like swipe_into_view [16:08] brendand: yes, that. [16:08] elopio, but i suppose it relies on autopilot being able to 'see' that list item? [16:09] brendand: look for the flickables module. It has some useful methods. And if the element is inside a flickable, it will have a method to do it too. [16:09] elopio, is there an easy way just to say 'flick to the end of the list'? [16:09] brendand: flickable.swipe_to_bottom [16:10] brendand: there's some magic to search for the item if it's not already loaded in the QML. [16:10] that needs an objectName for the item. [16:10] elopio, swipe to bottom should be sufficient [16:20] alesage: https://code.launchpad.net/~canonical-platform-qa/ueqa-code-proposals/indicator-sound-control_player/+merge/242805 please [16:20] elopio cool will look at [16:41] balloons, what do i need to do to coax ci tests to run here: https://code.launchpad.net/~brendan-donegan/ubuntu-clock-app/wait_for_bottomedgetip_visible/+merge/242792 [16:46] ubuntu-qa - review on https://code.launchpad.net/~brendan-donegan/reminders-app/test_add_notebook_must_append_it_to_list_swipe_to_bottom/+merge/242808 please [16:47] brendand: um, there should be only one notebook. [16:48] elopio, ? [16:48] elopio, there's a ton of them [16:49] brendand: each test should remove the notebooks it adds. [16:49] they are left because there's something wrong with the cleanup. [16:49] elopio, no the cleanup is fine [16:49] elopio, but there's a whole bunch of notebooks there [16:49] brendand: if the cleanaup was fine, there would be only one notebook. [16:50] elopio, it failed in ci though [16:50] elopio, where the test would only have run once [16:50] some tests hang or something. Jenkins kills the instance, and the notebooks are not deleted. I suppose. [16:50] brendand: this uses a real sandbox account. I have just cleaned the notebooks. [16:50] elopio, ah i see [16:51] brendand: I'm not sure your approach will work, because I'm not sure if the notebooks are sorted by date or by name. [16:51] elopio, do you not think it should swipe to the bottom just in case? [16:51] elopio, well the test is called 'append to list' [16:51] brendand: as it should be only one, it won't hurt swiping to bottom. [16:52] elopio, so if it doesn't append it then the test should fail [16:52] brendand: right, I named that but I don't know if it goes to the tail. [16:52] elopio, it does. obviously i tested the fix :P [16:52] ok then. I had written it for only two notebooks. [16:52] So +1. [16:53] elopio, i can see now it's not absolutely necessary, but it would prevent failures in the case where the sandbox account has become corrupted [16:54] brendand: yes. Agree. If you could keep an eye on it maybe we can improve the cleanup somehow. [16:56] elopio, btw do you know what i need to do to persuade the ci tests to run on my MPs? [16:56] brendand: I think you need to be part of a team. [16:57] not sure which one. balloons or cihelp should know. [16:57] elopio, maybe it's the team thomi just removed us all from :P ! [16:57] elopio I've added "request info" type comments; don't want to torture you but wanted some info :) [16:59] brendand: that's probably a ci help question on #ubuntu-ci-eng [17:00] alesage: the headset test is here: https://code.launchpad.net/~canonical-platform-qa/ueqa-code-proposals/indicator-sound-high_volume_warning/+merge/242809 [17:00] we listed two during our meeting. [17:00] the problem with that one is that I'm not yet able to make my prototype work. [17:01] alesage: answering your questions in the MP... [17:01] brendand, you need to be on the core apps qa team or the development team for the core app [17:02] balloons, who can add me? [17:02] balloons, thomi just removed canonical-platform-qa from it [17:02] I can and yes, that was the intent [17:02] balloons, so i guess we need to sign up individually [17:02] I can add individually sure [17:06] brendand, you are a core apps drivers now [17:07] balloons, so i just wait now? [17:07] elopio, did you see my comment on the location prompt merge? [17:07] brendand, which mp? [17:08] elopio, i can make it a fixture but i can't invest the time to put it in location service [17:08] elopio, since that's the rabbit hole of cmake and packaging [17:08] elopio, i did that once for url-dispatcher and i'd rather chew off my arm than do it again :) [17:09] brendand, that said, https://launchpad.net/~ubuntu-touch-coreapps-test-writers/+members shows the platform team is a part [17:09] brendand: that's the work man. I'm ok with you filing a bug and putting a comment on your code that it needs to be moved. [17:09] but we need a way to give that bug a high priority. [17:09] elopio, okay but who's going to address the bug? [17:09] it's going to be needed by the scopes tests too. [17:10] elopio, so for now make it a fixture and it can land? [17:11] the clock, the weather, the rss, the camera, many webapps will be location-aware. [17:11] brendand: yes, but lets agree that the next time it's needed, we won't copy it around. [17:11] we'll find a place for it, and the only way for getting managers to assign some time for it is to leave the test failing. [17:13] brendand: the url dispatcher is a case of success. We simplified the tests in dialer, messaging, unity and UX thanks to that one. Chewing your arm is not as good :) [17:13] elopio, we don't have to land it right now if you think it will expidite the 'proper' fix [17:13] elopio, heh - you didn't have to do it ! [17:13] actually, I think the dialer is not yet updated. [17:14] brendand: no, lets land it. Please file the bug. Then next time we hit it, we will have the code working, and good evidence of why it needs to be moved. [17:15] brendand: I know! You did it and I appreciate that. It should have been ted. That's the kind of thing we are documenting this week, an in theory all the managers have agreed to implement test helpers in their projects. [17:16] brendand: in your bug, please mention that when the fixture is moved it will need a test. [17:17] something like a simple qml app that only requests location, to check when the fixture is enabled and disabled. [17:17] brendand: oh, and one more detail. Please don't use print. Use the logger. [17:18] elopio, sure [17:18] brendand, jenkins is running for your stuff now [17:18] should be automagic [17:18] brendand: well, two details. Also please be consistent using the single and double quotes for strings. [17:18] for consistency with the rest of projects, use ' [18:35] elopio, here's this for you to tinker with slash fix https://code.launchpad.net/~ueqa-projects-team/ueqa-code-proposals/indicator-network/+merge/242826 [18:35] elopio I'm going to try the mac80211_hwsim hack now [18:35] cool. thanks alesage === geochr_ is now known as geochr === geochr is now known as geochr_ === geochr_ is now known as geochr [19:39] hallo doug5 [19:39] balloons, hello! [19:41] doug5, how are you? [19:43] balloons, great! you? [19:43] balloons, I'm actually trying to understand an error in calendar app [19:44] http://paste.ubuntu.com/9237601/ [19:45] ever seen? Otherwise I'll contact mihir [19:45] I'm doing well [19:45] * balloons looks [19:46] doug5, ahh yes that's a common error [19:46] hence the link given :-) http://developer.ubuntu.com/api/devel/ubuntu-14.10/python/autopilot/faq/troubleshooting.html [19:47] anyways, specific to that error, let's look at the test [19:47] basically it means the test didn't find the object. That most likely means a timing issue. You tried to get the object before it existed [19:49] but it's the fresh checked out source code [19:50] strange [19:54] doug5, yea, that simply means it probably wasn't accounted for when the test was created [19:54] * balloons blames the original author.. which is of course balloons [19:56] anyways, let's fix it. Do you understand why the error is happening? [19:58] * balloons runs test to have a look [20:00] so actually doug5 the first thing I do to troubleshoot if it's a timing thing (which I assumed it was) is to add a sleep before the failure. In this case that did not help. Instead, it's likely the object itself is no longer present, or has been renamed [20:01] balloons, yes that's my guess to [20:02] doug5, so if I manually run calendar I can see the UI selector for calendar is indeed missing [20:02] I think we can remove this check [20:02] no? [20:03] I think it's an open question for the devs, but yes for now we can certainly remove it [20:03] with it removed, everything passes and you should be able to carry on with the other checks [20:04] I noticed I cannot also add the event [20:04] file:///home/acerisara/Projects/ubuntu-calendar-app/NewEvent.qml:167: TypeError: Cannot read property 'collectionId' of undefined [20:09] doug5, you are right.. seems quite odd [20:09] I know they changed that screen around; I wonder if we are missing a depends or something [20:09] it seems rather odd it would be broke in trunk like that [20:13] * balloons installs all listed depends [20:13] doug5, that's the issue, things look correct for me now, including the calendar [20:13] doug5, I installed qmlscene qtdeclarative5-quicklayouts-plugin qtdeclarative5-ubuntu-ui-toolkit-plugin libqt5organizer5 qtdeclarative5-qtorganizer-plugin qtorganizer5-eds qtdeclarative5-ubuntu-syncmonitor0.1 libqt5contacts5 qtdeclarative5-qtcontacts-plugin qtcontact5-galera [20:13] that came from the debian/control file [20:16] let me try === roadmr is now known as roadmr_afk [20:18] balloons, yes...way better indeed :) thx [20:19] doug5, lol, see trunk wasn't broken after all.. just us :-) [20:20] at least we still have our sanity! [20:20] balloons, that's an excellent new actually :D [20:20] *news [20:37] veebers: alesage: pushed the changes. Do you want to take a look or should I merge to trunk? [20:37] elopio, PUSH [20:39] veebers: alesage: thomi: merged. === roadmr_afk is now known as roadmr [21:30] balloons, ok, changes are done :) tomorrow I will review and create the mr